new england

The
journal of medicine
established in 1812 August 5, 2021 vol. 385  no. 6

CRISPR-Cas9 In Vivo Gene Editing for Transthyretin Amyloidosis

Julian D. Gillmore, M.D., Ph.D., Ed Gane, M.B., Ch.B., Jorg Taubel, M.D., Justin Kao, M.B., Ch.B.,
Marianna Fontana, M.D., Ph.D., Michael L. Maitland, M.D., Ph.D., Jessica Seitzer, B.S., Daniel O’Connell, Ph.D.,
Kathryn R. Walsh, Ph.D., Kristy Wood, Ph.D., Jonathan Phillips, Ph.D., Yuanxin Xu, M.D., Ph.D., Adam Amaral, B.A.,
Adam P. Boyd, Ph.D., Jeffrey E. Cehelsky, M.B.A., Mark D. McKee, M.D., Andrew Schiermeier, Ph.D.,
Olivier Harari, M.B., B.Chir., Ph.D., Andrew Murphy, Ph.D., Christos A. Kyratsous, Ph.D., Brian Zambrowicz, Ph.D.,
Randy Soltys, Ph.D., David E. Gutstein, M.D., John Leonard, M.D., Laura Sepp‑Lorenzino, Ph.D.,
and David Lebwohl, M.D.​​

a bs t r ac t

BACKGROUND
Transthyretin amyloidosis, also called ATTR amyloidosis, is a life-threatening From the National Amyloidosis Centre,
Division of Medicine, University College
disease characterized by progressive accumulation of misfolded transthyretin
London, Royal Free Hospital (J.D.G.,
(TTR) protein in tissues, predominantly the nerves and heart. NTLA-2001 is an in M.F.) and Richmond Pharmacology, St.
vivo gene-editing therapeutic agent that is designed to treat ATTR amyloidosis by George’s University of London (J.T.) —
both in London; New Zealand Clinical
reducing the concentration of TTR in serum. It is based on the clustered regu-
Research (E.G.), University of Auckland
larly interspaced short palindromic repeats and associated Cas9 endonuclease (E.G.), and the Department of Neurolo-
(CRISPR-Cas9) system and comprises a lipid nanoparticle encapsulating messenger gy, Auckland City Hospital (J.K.) — all in
Auckland, New Zealand; Intellia Thera-
RNA for Cas9 protein and a single guide RNA targeting TTR.
peutics, Cambridge, MA (M.L.M., J.S.,
D.O., K.R.W., K.W., J.P., Y.X., A.A., A.P.B.,
METHODS J.E.C., M.D.M., A.S., J.L., L.S.-L., D.L.);
After conducting preclinical in vitro and in vivo studies, we evaluated the safety and Regeneron Pharmaceuticals, Tarry-
and pharmacodynamic effects of single escalating doses of NTLA-2001 in six pa- town, NY (O.H., A.M., C.A.K., B.Z., R.S.,
D.E.G.). Address reprint requests to Prof.
tients with hereditary ATTR amyloidosis with polyneuropathy, three in each of the Gillmore at the Centre for Amyloidosis
two initial dose groups (0.1 mg per kilogram and 0.3 mg per kilogram), within an and Acute Phase Proteins, Division of
ongoing phase 1 clinical study. Medicine, University College London,
Royal Free Hospital, Rowland Hill St.,
London NW3 2PF, United Kingdom, or at
RESULTS
­j​.­gillmore@​­ucl​.­ac​.­uk.
Preclinical studies showed durable knockout of TTR after a single dose. Serial as-
This article was published on June 26,
sessments of safety during the first 28 days after infusion in patients revealed few
2021, at NEJM.org.
adverse events, and those that did occur were mild in grade. Dose-dependent
N Engl J Med 2021;385:493-502.
pharmacodynamic effects were observed. At day 28, the mean reduction from
DOI: 10.1056/NEJMoa2107454
baseline in serum TTR protein concentration was 52% (range, 47 to 56) in the Copyright © 2021 Massachusetts Medical Society.
group that received a dose of 0.1 mg per kilogram and was 87% (range, 80 to 96)
in the group that received a dose of 0.3 mg per kilogram.
CONCLUSIONS
In a small group of patients with hereditary ATTR amyloidosis with polyneuropa-
thy, administration of NTLA-2001 was associated with only mild adverse events
and led to decreases in serum TTR protein concentrations through targeted
knockout of TTR. (Funded by Intellia Therapeutics and Regeneron Pharmaceuti-
cals; ClinicalTrials.gov number, NCT04601051.)

n engl j med 385;6  nejm.org  August 5, 2021 493

The New England Journal of Medicine
Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l
T
ransthyretin amyloidosis, also
called ATTR amyloidosis, is a progressive
fatal disease characterized by accumula-
tion in tissues of amyloid fibrils composed of
misfolded transthyretin (TTR) protein.1,2 ATTR
A Quick Take
is available at amyloidosis may be acquired; referred to as wild-
NEJM.org type ATTR amyloidosis, this form of ATTR amy-
loidosis is an increasingly recognized cause of
cardiomyopathy and heart failure.2 In rarer cases,
ATTR amyloidosis is hereditary (known as vari-
ant or hereditary ATTR [hATTR] amyloidosis);
hATTR amyloidosis can be triggered by more
than 100 different pathogenic mutations in TTR.3
The hereditary form of ATTR amyloidosis is
thought to be present in approximately 50,000
persons worldwide4,5; it has an autosomal domi-
nant pattern of inheritance and a clinical pheno-
type dominated by amyloid polyneuropathy or
cardiomyopathy, with most patients having a
combination of the two.6 After the onset of symp-
toms, ATTR amyloidosis is progressive, culmi-
nating in death within a median of 2 to 6 years
after diagnosis in patients with amyloid cardio-
myopathy7 and 4 to 17 years after symptom onset
in patients with amyloid polyneuropathy in the
absence of cardiomyopathy.8
Current therapeutic strategies for ATTR amy-
loidosis rely on reducing ongoing amyloid for-
mation through stabilization of the tetrameric
form of TTR (with diflunisal or tafamidis)9,10 or
through inhibition of TTR protein synthesis
(with inotersen or patisiran) by means of degra-
dation of TTR messenger RNA (mRNA).11,12 Such
treatments produce symptom relief and functional
improvement and prolong survival12-14 but are lim-
ited by the requirement for long-term adminis-
tration to maintain TTR knockdown. In the case
of patisiran, long-term treatment results in con-
tinued exposure to premedication with glucocor-
ticoids and antihistamines.15 In addition, patients
receiving TTR-stabilizing agents have disease pro-
gression.16 Inotersen is associated with serious
side effects, including glomerulonephritis and
decreased platelet counts.17 More extensive TTR
knockdown is associated with greater improve-
ment in neuropathic end points in patients with
hATTR polyneuropathy.12 Enhancements in TTR
reduction, including sustained knockdown, may
translate into improved outcomes for patients with
ATTR amyloidosis. A potential alternative to
mRNA targeting-based gene silencing is use of the
494 n engl j med 385;6  nejm.org  August 5, 2021
The New England Journal of Medicine
Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
clustered regularly interspaced short palindrom-
ic repeats and associated Cas9 endonuclease
(CRISPR-Cas9) system to achieve in vivo gene
editing.18-21 As a monogenic disease, ATTR amy-
loidosis represents an ideal target for the appli-
cation of CRISPR-Cas9–mediated in vivo gene
editing. The limited and specific normal func-
tion of TTR in thyroxine and vitamin A trans-
port22 means that knockdown has only limited
additional physiological effects23; in addition,
circulating TTR is produced almost entirely (>99%)
within the liver,2 for which established targeting
systems such as lipid nanoparticles are available.
NTLA-2001 is a new CRISPR-Cas9–based in
vivo gene-editing therapy, administered by intra-
venous infusion, that is intended to edit TTR in
hepatocytes, leading to a decrease in the produc-
tion of both wild-type and mutant TTR after a
single administration. In mouse models and non-
human primates (cynomolgus monkeys), single
doses resulted in durable reductions in serum
TTR protein of 95% or greater24,25 and therefore
provide potentially greater TTR knockdown
than currently available therapies.
NTLA-2001 consists of a proprietary lipid
nanoparticle (LNP) delivery system with liver
tropism, carrying a single guide RNA (sgRNA)
that targets human TTR and a human-codon–
optimized mRNA sequence of Streptococcus pyo-
genes Cas9 protein (Fig. 1). LNPs have previously
been used in vivo for liver-targeted delivery of a
variety of therapeutic RNA cargoes, such as small
interfering RNA and mRNA.26,27 For NTLA-2001,
the LNP formulation was developed for genome
editing with regard to lipid composition and RNA
cargo in order to enable efficient delivery to the
liver in a variety of preclinical models. As de-
scribed previously, on polyethylene glycol–lipid
diffusion from this LNP, plasma apolipoprotein
E binds to (opsonizes) the LNP surface in circu-
lation; the LNP is then actively endocytosed by
hepatocytes through the low-density lipoprotein
receptor.27 Given that the liver is the almost exclu-
sive site of TTR manufacture,2 this liver-targeting
delivery system should maximize efficacy while
minimizing systemic toxic effects.
Here, we report interim data from an ongoing
clinical study evaluating single ascending doses
of NTLA-2001 for TTR editing and knockout in
the treatment of patients with hATTR amyloido-
sis with polyneuropathy.
 CRISPR-C as 9 for Tr ansthyretin Amyloidosis
Me thods
Preclinical Studies
An sgRNA targeting the TTR sequence
AAAGGCUGCUGAUGACACCU (human genome
build hg38, chromosome 18: 31592987–31593007)
was selected for efficient knockout and specific-
ity after a comprehensive off-target characteriza-
tion workflow that applied a combination of
both computational modeling and empirical
approaches. To select for a high therapeutic in-
dex (i.e., the ratio of on-target to off-target edit-
ing), we performed genomewide assays and tar-
geted sequencing to identify and verify candidate
sgRNA off-target sites. The in vitro dose–response
and gene-editing potency of NTLA-2001 were
assessed in primary cell cultures of human he-
patocytes. Genomic loci with the potential for
off-target editing were found with the use of
complementary computational and laboratory-
based approaches (Cas-OFFinder,28 GUIDE-seq,29
and SITE-Seq30). Candidate loci were validated
for the detection of off-target insertions and
deletions (indels) with the use of next-generation
sequencing after NTLA-2001 treatment of pri-
mary human hepatocytes at concentrations up to
27 times as high as concentrations that achieved
greater than 90% reduction in TTR protein (EC90).
Additional details of these studies are provided
in the Supplementary Appendix, available with
the full text of this article at NEJM.org.
Transgenic mouse models with either wild-
type or mutant (p.V50M) human TTR were used
for initial proof-of-mechanism studies of NTLA-
2001 in which we evaluated TTR editing and re-
ductions in circulating TTR protein levels.24 The
cynomolgus monkey was selected as a relevant
nonclinical animal model for the evaluation of
pharmacologic, toxicologic, pharmacokinetic,
and metabolic properties, including TTR editing,
reductions in circulating TTR protein levels, and
serum concentrations of LNP and cargo compo-
nents over time.25
Clinical Study Oversight
The clinical study is sponsored by Intellia Thera-
peutics and Regeneron Pharmaceuticals; the spon-
sors designed the protocol, available at NEJM.org.
Study oversight was provided by an independent
data and safety monitoring committee. Data were
collected by study investigators in New Zealand
n engl j med 385;6  nejm.org  August 5, 2021
The New England Journal of Medicine
Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
and the United Kingdom and were analyzed by
the sponsors. A medical writer prepared the first
draft of the manuscript, under direction from
the authors and with funding from the sponsors.
The authors had access to the data and collabo-
rated in the preparation and revisions of manu-
script drafts before and after submission. The
investigators and sponsors vouch for the accu-
racy and completeness of the data and for the
fidelity of the study to the protocol. The protocol
was reviewed by national and institutional ethics
and regulatory bodies, including expert commit-
tees for the assessment of new studies of gene
therapy, such as gene editing (Gene Therapy Ad-
visory Committee of the Ministry of Health, New
Zealand, and Gene Therapy Advisory Committee
of the Medicines and Healthcare Products Regu-
latory Agency, United Kingdom). The study has
been conducted in accordance with the Declara-
tion of Helsinki and International Council for
Harmonisation Good Clinical Practice guide-
lines, and all patients provided written informed
consent.
Clinical Study Design and Eligibility
We report the results in two initial dose groups
from part 1 of a two-part, global, phase 1, open-
label, multicenter study. Patients were treated with
a single dose of NTLA-2001 consisting of a total
RNA dose of 0.1 mg per kilogram of body weight
or 0.3 mg per kilogram of body weight adminis-
tered intravenously between November 2020 and
April 2021. Key eligibility criteria for part 1 of the
study included an age of 18 to 80 years, a diag-
nosis of polyneuropathy due to hATTR amyloi-
dosis (with or without cardiomyopathy), a body
weight of 50 to 90 kg at the screening visit, and
a lack of access to approved treatments for ATTR
amyloidosis. Patients with non-ATTR amyloido-
sis, known leptomeningeal ATTR amyloidosis,
or a history of receipt of RNA-silencing therapy
were excluded. Previous use of TTR stabilizers
was permitted with a washout period (3 days for
diflunisal).
Clinical Study Treatment
Safety studies in cynomolgus monkeys were used
to determine the no-observed-adverse-effect level
(NOAEL), which was found to be a single admin-
istration of 3 mg per kilogram infused intrave-
nously, equivalent to a dose of 1 mg per kilogram
495
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Intravenous Infusion of NTLA-2001

ApoE opsonization
of LNP in circulation
NTLA-2001

TTR-specific
sgRNA

Complementary
sequence to ApoE protein
TTR gene

Streptococcus Rapid distribution to liver

pyogenes (Spy) through hepatic artery
Cas9 mRNA
Lipid nanoparticle

B NTLA-2001 LNP Uptake in Hepatocytes C Cleavage of DNA at TTR Gene Sequence by Cas9

HEPATOCYTE

CAPILLARY
ApoE protein NUCLEUS

Space of Disse
LDL receptor
HEPATOCYTE sgRNA
Endocytosis of LNP
Cas9

Endosome
Genomic DNA HNH
Complementary
3' 5'
Breakdown of LNP and
release of Cas9 mRNA and
5' PAM 3'
TTR-specific sgRNA into
cytoplasm Noncomplementary
RuvC

Ribonucleoprotein induces Targeted DNA

unwinding of chromosomal cleavage
Translation of Cas9 mRNA DNA, then binds to DNA at
TTR gene, inducing
targeted DNA cleavage 3' 5'

+ 5' 3'
TTR-specific
sgRNA Spy Cas9
endonuclease

Cas9–sgRNA ribonucleoprotein
Scaffold region Endogenous DNA repair
Cas9 through nonhomologous end
sgRNA
Target-specific joining results in introduc-
20-nt sequence 5' tion of indels into TTR gene,
leading to frameshift
mutations that prevent
production of functional TTR
NUCLEUS Cas9–sgRNA ribonucleoprotein protein
enters nucleus

496 n engl j med 385;6 nejm.org August 5, 2021

The New England Journal of Medicine

Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 CRISPR-C as 9 for Tr ansthyretin Amyloidosis
Figure 1 (facing page). Mechanism of Action of NTLA-
2001.
Panel A shows the primary components of NTLA-
2001. The carrier system for NTLA-2001 is a lipid
nanoparticle (LNP). The LNP is based on a proprietary
ionizable lipid, combined with a phospholipid, a peg­
ylated lipid (molecular weight of polyethylene glycol,
2000 Da), and cholesterol, formulated in an aqueous
buffer for intravenous administration. The active com-
ponents of NTLA-2001 are a human-optimized mes-
senger RNA (mRNA) molecule encoding Streptococcus
pyogenes (Spy) Cas9 protein (an approximately
4400-nucleotide sequence with a molecular weight
of approximately 1.5 MDa) and a single guide RNA
(sgRNA) molecule (molecular weight of approximately
35 kDa) specific to the human gene encoding trans-
thyretin (TTR). These components form the cargo of
the LNP for drug administration. After intravenous ad-
ministration of NTLA-2001 and entry into the circula-
tion, the LNP is opsonized by apolipoprotein E (ApoE)
and transported through the systemic circulation di-
rectly into the liver, where it is preferentially distributed.
Panel B shows transport of the NTLA-2001 LNP into
the capillaries of the hepatic sinusoids inside the liver.
As with other clinically approved LNPs,27 NTLA-2001
is then expected to undergo uptake by the low-density
lipoprotein (LDL) receptor expressed on the surface
of the hepatocytes, followed by endocytosis and endo-
some formation. After breakdown of the LNP and dis-
ruption of the endosomal membrane, the active com-
ponents (the TTR-specific sgRNA and the mRNA
encoding Cas9) are released into the cytoplasm. The
Cas9 mRNA molecule is translated through the native
ribosomal process, producing the Cas9 endonuclease
enzyme. The TTR-specific sgRNA interacts with the
Cas9 endonuclease, forming a clustered regularly in-
terspaced short palindromic repeats (CRISPR)–Cas9
ribonucleoprotein complex. Panel C shows that the
Cas9 ribonucleoprotein complex is targeted for nuclear
import and enters the nucleus, where it recognizes
the protospacer-adjacent motif (PAM) on the noncom-
plementary DNA strand in TTR. A target-specific
20-nucleotide sequence at the 5′ end of the sgRNA
binds to the DNA double helix at the target site, allow-
ing the CRISPR-Cas9 complex to unwind the helix and
access the target gene. Cas9 undergoes a series of
conformational changes and nuclease domain activa-
tion (HNH and RuvC domains), resulting in DNA
cleavage that is precisely targeted to the TTR sequence,
as defined by the sgRNA complementary sequence.
Endogenous DNA-repair mechanisms ligate the ends
of the cut, potentially introducing insertions or dele-
tions of bases (indels). The generation of an indel may
result in the reduction of functional target-gene mRNA
levels as a result of missense or nonsense mutations
decreasing the amount of full-length mRNA, ultimate-
ly resulting in decreased levels of the target protein.
Indels that result in abrogated production of the tar-
get protein, in this case TTR, are termed knockout
mutations.
n engl j med 385;6  nejm.org  August 5, 2021
The New England Journal of Medicine
Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
in humans. In accordance with allometric scaling
based on total body-surface area and application
of a safety factor of 10, the maximum recom-
mended starting dose of NTLA-2001 for this study
was 0.1 mg per kilogram. To mitigate against
potential proinflammatory effects of intravenous
LNP infusions, patients received glucocorticoid
and histamine receptor type 1 and type 2 block-
ade before infusion.
Assessments of Clinical Outcomes
Patients were monitored for assessment of ad-
verse events and laboratory findings. Serum sam-
ples were obtained at baseline and at weeks 1, 2,
and 4 for analysis of TTR protein levels with a
validated enzyme-linked immunosorbent assay.
Evaluations of safety and therapeutic-activity out-
comes are planned for 24 months after NTLA-
2001 infusion. In accordance with the require-
ments of the regulatory authorities that approved
the current protocol, longer-term safety follow-
up is planned under a separate program currently
in development.
Statistical Analysis
Descriptive analyses only were planned. Measure-
ments of serum TTR protein levels at baseline
were compared with those at day 28 and are pre-
sented as the mean percentage change and range.
R e sult s
Preclinical Studies of NTLA-2001
Preclinical development of NTLA-2001 (Fig. 1)
included computational modeling and in vitro
studies to ensure on-target gene editing (see the
Supplementary Appendix). In primary human he-
patocytes, NTLA-2001 was highly potent (EC50,
0.05 to 0.15 nmol per liter; EC90, 0.17 to 0.67
nmol per liter) and produced saturating levels of
TTR editing (≥93.7%), resulting in 91% or great-
er reductions in TTR mRNA expression and 95%
or greater reductions in TTR protein production
(Fig. 2). Data from next-generation sequencing
showed that NTLA-2001 induced knockout of TTR.
In off-target editing assays, among all the po-
tential off-target loci identified by Cas-OFFinder,
GUIDE-seq, and SITE-Seq, seven loci were iden-
tified as possible editing sites, all of which were
located in noncoding regions (Fig. S1 in the Sup-
plementary Appendix). For each of these sites,
no evidence of off-target editing was found when
497
 The n e w e ng l a n d j o u r na l
TTR gene editing TTR mRNA expression TTR protein
production
100
90
80
70
Percentage
60
50
40
30
20
10
0
0.007 0.021 0.063 0.191 0.574 1.724 5.172 15.517
sgRNA Concentration (nmol/liter)
Figure 2. In Vitro Evaluations of the Potency of NTLA-2001.
Shown is the relationship between increasing concentrations of sgRNA and
the consequent percentages of TTR editing, as well as TTR mRNA expres-
sion and TTR protein production in a single lot of primary human hepato-
cytes. The primary indel patterns were a single-nucleotide deletion or in-
sertion at the cut site, inducing a frameshift mutation (data not shown).
primary human hepatocytes were treated with
concentrations of NTLA-2001 up to 3 times as
high as the EC90 (Fig. S2).
Studies in transgenic mice revealed a dose-
dependent and durable effect of NTLA-2001.
Editing of TTR reduced circulating serum TTR
protein levels, which reached a nadir by 4 weeks
after receipt of the dose and were still maxi-
mally suppressed at 12 months of observation.24
After resection of two thirds of the liver and
subsequent full-liver regeneration, the gene-edit-
ing percentage and corresponding protein levels
were unchanged, findings that supported the
permanent nature of the edit (Fig. S7).
Studies in cynomolgus monkeys showed rap-
id initial distribution and clearance of the LNP
components (Table S3 and Fig. S8). In addition,
a single dose of Cyn-LNP (the nonhuman primate
surrogate of NTLA-2001) at 3 or 6 mg per kilogram
was associated with a maximum gene-editing
percentage of 73% in whole liver and near-com-
plete (>94%) reduction in serum TTR protein
that was sustained over a period of 12 months
(Fig. 3A). Editing of TTR was confirmed by next-
generation sequencing analysis of hepatic tissue
(Fig. 3B).25
Thus, preclinical studies in the mouse and
cynomolgus monkey showed that a single dose of
NTLA-2001 or its surrogate Cyn-LNP resulted in
durable TTR editing and near-complete elimina-
498 n engl j med 385;6  nejm.org  August 5, 2021
The New England Journal of Medicine
Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
tion of serum TTR protein expression at doses
associated with no adverse effects. To evaluate the
ability of NTLA-2001 to reduce serum expression
of TTR protein after a single intravenous infu-
sion in humans, we initiated an open-label, single-
dose, proof-of-concept study involving patients
who had hATTR amyloidosis with polyneuropathy.
Patients
At one study site (Auckland, New Zealand), three
patients underwent screening, of whom two were
found to be eligible and were enrolled. One par-
ticipant had a body weight that was above the
upper limit allowed by the study protocol at that
time. At the other study site (London, United King-
dom), four patients underwent screening, all of
whom were found to be eligible and were enrolled.
The patients were 46 to 64 years of age, and four
of the six patients were men; the body weight
ranged from 70 to 90 kg. Three patients had a
p.T80A mutation, two a p.S97Y mutation, and one
a p.H110D mutation. Three patients had received
no previous therapy, and three had previously
received diflunisal. All six patients had sensory
polyneuropathy in the absence of motor symp-
toms (polyneuropathy disability score of 1) and
a New York Heart Association heart failure class
of I. The level of N-terminal pro–B-type natriuretic
peptide ranged from 50 to 596 ng per liter.
Safety and Side-Effect Profile
NTLA-2001 treatment was completed in all the
patients without interruption of the infusion. No
protocol-specified stopping events were observed.
Adverse events that occurred during or after
treatment, all of which were mild (grade 1) in
severity, were reported in three of the six pa-
tients. One patient had an adverse event of spe-
cial interest (a grade 1 infusion-related reaction;
see Table S4). No serious adverse events were
observed. Increased d-dimer levels were ob-
served 4 to 24 hours after infusion in five of six
patients; the elevations were lower than those
observed at the NOAEL dose in nonhuman pri-
mates. The values returned to baseline in all six
patients by day 7. Coagulation measures (acti-
vated partial thromboplastin time and pro-
thrombin time) remained within 1.2 times the
upper limit of the reference ranges, and fibrino-
gen levels and platelet counts remained above
the lower limit of the reference ranges; liver-
 CRISPR-C as 9 for Tr ansthyretin Amyloidosis

function measures (aspartate aminotransferase

A
and alanine aminotransferase levels) remained Control 1.5 mg/kg 3.0 mg/kg 6.0 mg/kg
within normal limits (Fig. S11). (N=3) (N=3) (N=3) (N=3)
120
TTR Protein Reduction 100
To determine the pharmacodynamic effects of

TTR Concentration
(% of baseline)
80
NTLA-2001, concentrations of TTR in serum were
evaluated. Reductions from baseline in the se- 60
rum TTR protein concentration were observed 40
by day 14 and deepened by day 28 (Fig. 4A and
20
4B). At day 28, NTLA-2001 was associated with
mean TTR reductions of 52% in the group that 0
received a dose of 0.1 mg per kilogram and 87% 0 40 80 120 160 200 240 280 320 360

in the group that received 0.3 mg per kilogram Day

(Fig. 4C). The effect was dose-dependent, with
greater reductions in TTR concentration among
patients who received a higher dose of NTLA-
2001. In addition, the effect of NTLA-2001 was
reproducible across patients at each dose level,
with reductions at day 28 ranging from 47 to 56%
in the lower-dose group and from 80 to 96% in
the higher-dose group (Fig. 4A and 4B).
Discussion
We report evidence of CRISPR-Cas9–based in vivo
gene editing in humans. Systemic administration
of NTLA-2001 to six patients with hATTR amy-
loidosis with polyneuropathy was associated in
each case with sustained reductions in the serum
TTR protein concentration. NTLA-2001 treatment
was associated with a dose-dependent effect. At
day 28, the time at which the drug effect had
reached its permanent nadir in preclinical stud-
ies, the mean reduction from baseline in serum
TTR protein concentration was 52% in the group
that received the lower dose (0.1 mg per kilo-
gram) and was 87% in the group that received
the higher dose (0.3 mg per kilogram). NTLA-2001
treatment was associated with adverse events of
only mild severity.
These data represent interim results from the
first two dose groups in an ongoing dose-esca-
lation study. The results closely follow the pat-
tern observed in in vitro data from cell lines and
in vivo data on potency in animals, which showed
a deep and permanent reduction in serum TTR
protein concentrations with NTLA-2001, thus
providing evidence of the potential for in vivo
gene editing as a therapeutic strategy for the
treatment of hATTR amyloidosis. It is important
B
Indel
Indel Size Sequence Frequency
%
Wild type —
+1 98.9
>+1 1.03
Figure 3. In Vivo Pharmacologic Properties of Cyn-LNP, the Nonhuman
­Primate Surrogate of NTLA-2001.
Panel A shows mean reductions in the serum TTR protein concentration as
a percentage of the baseline concentration in cynomolgus monkeys (three
per dose group) that received Cyn-LNP intravenously at doses of 1.5, 3.0,
and 6.0 mg of total RNA per kilogram of body weight on day 0 and were fol-
lowed up for 367 days. A control group that received no treatment is shown
for comparison. I bars indicate standard deviations for the three animals in
each group. Panel B shows the results of next-generation sequencing after
administration of Cyn-LNP to cynomolgus monkeys. The sgRNA target se-
quence is indicated in blue next to the required PAM sequence, shown in
red. [G/A] represents a naturally occurring single-nucleotide polymorphism
in the cynomolgus monkeys used in the study. The nucleotide position of
indels relative to the cynomolgus monkey genome (build mf5, chromo-
some 18) are +1: 50681549–50681550. The primary indel pattern was a single-
nucleotide insertion at the cut site, inducing a frameshift mutation. An “N”
at the insertion site indicates a multinucleotide insertion (e.g., AA or AGG),
which in aggregate constituted 1.03% of all indels. The remaining fraction
comprised deletions of various lengths.
to note that this study involves a very small num-
ber of patients with limited follow-up to date.
Continued serial measurements of serum TTR
concentration in the patients reported here are
planned to confirm the durability of the effect.
Data from studies of RNA-targeting gene-
silencing agents have shown that observed re-
ductions in serum TTR protein translate into
meaningful clinical benefits relative to placebo.
These agents result in mean reductions from base-
line in serum TTR concentrations of approximate-

n engl j med 385;6  nejm.org  August 5, 2021 499

The New England Journal of Medicine
Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Change in Serum TTR Concentration in Patients Who Received 0.1 mg/kg Figure 4. Reductions from Baseline in Serum TTR
0 ­Protein Concentration after Infusion of NTLA-2001
in Humans.
−10
Panel A shows the percentage change from baseline in
−20
Change from Baseline (%)

total circulating serum TTR protein in the group of pa-

−30 tients who received an NTLA-2001 dose of 0.1 mg per
−40 kilogram. TTR protein was quantified by a validated
−47 enzyme-linked immunosorbent assay method in accor-
−50 −52
−56
dance with regulatory guidelines for biomarker method
−60
validation. Serum samples were measured once, and
−70 each sample was tested in duplicate. In accordance
−80 with good laboratory practice, no retesting was con-
ducted for successful assay runs. For each patient in
−90
the group that received a dose of 0.1 mg per kilogram,
−100 data are shown at postdose days 7, 14, and 28 for per-
0 7 14 21 28
centage reductions from the predose baseline value
Day (mean concentration from three sampling time points).
Panel B shows the percentage change from baseline
B Change in Serum TTR Concentration in Patients Who Received 0.3 mg/kg in total circulating serum TTR protein concentrations
0 in the group of patients who received an NTLA-2001
−10 dose of 0.3 mg per kilogram. The methods and analy-
sis were identical to those described in Panel A. Panel
−20
Change from Baseline (%)

C shows the mean (three per group) percentage reduc-

−30 tion from baseline in total circulating serum TTR pro-
−40 tein at day 28 for both dose groups. The I bars indicate
standard deviations.
−50
−60
−70
−80 −80 serial infusions. On the basis of data in animals,
−84
−90 NTLA-2001 may be able to produce nearly com-
−96
−100 plete and permanent knockdown of TTR expres-
0 7 14 21 28
sion with a single administration. As with cur-
Day rent standard-of-care agents, patients will receive
vitamin A supplementation in order to compen-
C Mean Reduction in Serum TTR Level at Day 28
sate for the loss of TTR, which has a normal
0
physiological role in vitamin A transport.31
−10
The potential for off-target gene editing with
−20
CRISPR-Cas systems has been raised as a concern
Mean Reduction (%)

−30
with regard to the use of these therapies in hu-
−40 mans.32 In primary human hepatocytes, therapeu-
−50 tic concentrations of NTLA-2001 showed no
−60 evidence of previously described off-target muta-
−70 genesis mechanisms.33,34 Computational model-
−80 ing, biochemical cell-free assays, and in vitro
−90 cellular assays were used to identify the most
−100 likely sites in the genome outside of TTR into
0.1 mg/kg 0.3 mg/kg
which NTLA-2001 might introduce off-target
Dose edits. Seven candidate sites for introduction of
indels were confirmed in assays with cell lines
in which supersaturating doses of NTLA-2001
ly 80% and show more favorable clinical effects were used. We assessed the therapeutic index for
in patients in whom lower concentrations of TTR NTLA-2001 in primary human hepatocytes by
protein are achieved. Maintenance of these reduc- determining the frequency of off-target edits at
tions with RNA-targeting agents requires routine concentrations of sgRNA that caused high de-

500 n engl j med 385;6  nejm.org  August 5, 2021

The New England Journal of Medicine

Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 CRISPR-C as 9 for Tr ansthyretin Amyloidosis

grees of on-target edits and reductions in TTR
protein production. In addition, we performed
long-range sequencing to detect structural vari-
ants and to determine the risk of unintended
genotoxicity or oncogenic transformation.35,36
Indels and DNA structural variants are natural
outcomes of double-stranded DNA break re-
pair.36 The DNA structural variants induced by
CRISPR-Cas9 genome editing were not random
but related to end-joining at the TTR on-target
site and are expected to be of low risk. The
changes detected in the primary human hepato-
cytes could predict the DNA structural variants
that may occur in vivo. Participants who volunteer
to receive NTLA-2001 therapy will need to undergo
long-term safety monitoring.
The CRISPR-Cas9 approach used for NTLA-2001
is modular and has the capacity to be adapted to
treat other diseases with simple replacement of
the sgRNA. Indeed, the in vivo gene-editing ap-
proach used in this study is currently being in-
vestigated for use in other diseases. Further
clinical programs involving CRISPR-Cas9–based
gene-editing strategies are planned by many in-
vestigators for a wide range of diseases; these
programs may make use of the potential not
only to knock out expression of harmful protein
products but also to insert genes to produce func-
tional proteins where mutations cause pathologic
deficiencies.
The study reported here is ongoing. Dose es-
calation continues with a goal of producing greater
reductions in serum TTR protein than are achieved
with available therapies, with anticipated benefi-
cial effects on disease progression, quality of life,
and mortality. Data from the initial groups of
patients in this study provide clinical proof of con-
cept for in vivo CRISPR-Cas9–mediated gene
editing as a therapeutic strategy.
Supported by Intellia Therapeutics and Regeneron Pharma-
ceuticals.
Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
A data sharing statement provided by the authors is available
with the full text of this article at NEJM.org.
We thank the patients who participated in this study and
their families; New Zealand Clinical Research and Richmond
Pharmacology for contract research assistance; Altasciences,
Charles River Laboratories, Precision for Medicine, PPD, and
QPS for assistance with the studies in cynomolgus monkeys,
enzyme-linked immunosorbent assay measurements of serum
transthyretin, and pharmacokinetic and biomarker tests in the
first-in-human study; Carri Boiselle, James Butler, David Cooke,
Tracy DiMezzo, Richard Duncan, Eva Essig, Noah Gardner, Bo
Han, Denise Hernandez, Tracy Jennings, Kellie Kolb, Rebecca
Lescarbeau, Reynald Lescarbeau, Nishit Patel, Austin Ricker,
Joseph Rissman, Matthew Roy, Philipp Schneggenburger, Palak
Sharma, and Samantha Soukamneuth from Intellia Therapeu-
tics for their dedicated investment in the development of NTLA-
2001; and Ben Caldwell of Arc, a division of Spirit Medical Com-
munications Group, for providing medical writing services.

References
1. Marcoux J, Mangione PP, Porcari R,
et al. A novel mechano-enzymatic cleav-
age mechanism underlies transthyretin
amyloidogenesis. EMBO Mol Med 2015;​7:​
1337-49.
2. Gertz MA, Benson MD, Dyck PJ, et al.
Diagnosis, prognosis, and therapy of
transthyretin amyloidosis. J Am Coll Car-
diol 2015;​66:​2451-66.
3. Ando Y, Coelho T, Berk JL, et al.
Guideline of transthyretin-related heredi-
tary amyloidosis for clinicians. Orphanet
J Rare Dis 2013;​8:​31.
4. Hawkins PN, Ando Y, Dispenzeri A,
Gonzalez-Duarte A, Adams D, Suhr OB.
Evolving landscape in the management of
transthyretin amyloidosis. Ann Med 2015;​
47:​625-38.
5. Schmidt HH, Waddington-Cruz M,
Botteman MF, et al. Estimating the global
prevalence of transthyretin familial amy-
loid polyneuropathy. Muscle Nerve 2018;​
57:​829-37.
6. Dohrn MF, Ihne S, Hegenbart U, et al.
Targeting transthyretin — mechanism-
based treatment approaches and future
perspectives in hereditary amyloidosis.
J Neurochem 2021;​156:​802-18.
7. Maurer MS, Bokhari S, Damy T, et al.
Expert consensus recommendations for
the suspicion and diagnosis of transthyre-
tin cardiac amyloidosis. Circ Heart Fail
2019;​12(9):​e006075.
8. Merlini G, Coelho T, Waddington
Cruz M, Li H, Stewart M, Ebede B. Evalu-
ation of mortality during long-term treat-
ment with tafamidis for transthyretin
amyloidosis with polyneuropathy: clinical
trial results up to 8.5 years. Neurol Ther
2020;​9:​105-15.
9. Maurer MS, Schwartz JH, Gundapa-
neni B, et al. Tafamidis treatment for pa-
tients with transthyretin amyloid cardio-
myopathy. N Engl J Med 2018;​379:​1007-16.
10. Berk JL, Suhr OB, Obici L, et al. Re-
purposing diflunisal for familial amyloid
polyneuropathy: a randomized clinical
trial. JAMA 2013;​310:​2658-67.
11. Benson MD, Waddington-Cruz M,
Berk JL, et al. Inotersen treatment for pa-
tients with hereditary transthyretin amy-
loidosis. N Engl J Med 2018;​379:​22-31.
12. Adams D, Gonzalez-Duarte A,
O’Riordan WD, et al. Patisiran, an RNAi
therapeutic, for hereditary transthyretin
amyloidosis. N Engl J Med 2018;​379:​11-21.
13. Adams D, Polydefkis M, González-
Duarte A, et al. Long-term safety and
efficacy of patisiran for hereditary trans-
thyretin-mediated amyloidosis with poly-
neuropathy: 12-month results of an open-
label extension study. Lancet Neurol 2021;​
20:​49-59.
14. Solomon SD, Adams D, Kristen A, et al.
Effects of patisiran, an RNA interference
therapeutic, on cardiac parameters in pa-
tients with hereditary transthyretin-medi-
ated amyloidosis. Circulation 2019;​ 139:​
431-43.
15. Urits I, Swanson D, Swett MC, et al.
A review of patisiran (ONPATTRO) for the
treatment of polyneuropathy in people
with hereditary transthyretin amyloido-
sis. Neurol Ther 2020;​9:​301-15.
16. Lozeron P, Théaudin M, Mincheva Z,
Ducot B, Lacroix C, Adams D. Effect on
disability and safety of tafamidis in late
onset of Met30 transthyretin familial
amyloid polyneuropathy. Eur J Neurol
2013;​20:​1539-45.
17. Gertz MA, Scheinberg M, Wadding-
ton-Cruz M, et al. Inotersen for the treat-
ment of adults with polyneuropathy
caused by hereditary transthyretin-medi-

n engl j med 385;6  nejm.org  August 5, 2021 501

The New England Journal of Medicine
Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 CRISPR-C as 9 for Tr ansthyretin Amyloidosis

ated amyloidosis. Expert Rev Clin Phar-
macol 2019;​12:​701-11.
18. Jinek M, Chylinski K, Fonfara I, Hauer
M, Doudna JA, Charpentier E. A program-
mable dual-RNA-guided DNA endonucle-
ase in adaptive bacterial immunity. Science
2012;​337:​816-21.
19. Hsu PD, Lander ES, Zhang F. Develop-
ment and applications of CRISPR-Cas9 for
genome engineering. Cell 2014;​157:​1262-
78.
20. Dasgupta I, Flotte TR, Keeler AM.
CRISPR/Cas-dependent and nuclease-free
in vivo therapeutic gene editing. Hum
Gene Ther 2021;​32:​275-93.
21. Li H, Yang Y, Hong W, Huang M, Wu
M, Zhao X. Applications of genome edit-
ing technology in the targeted therapy of
human diseases: mechanisms, advances
and prospects. Signal Transduct Target
Ther 2020;​5:​1.
22. Power DM, Elias NP, Richardson SJ,
Mendes J, Soares CM, Santos CR. Evolu-
tion of the thyroid hormone-binding pro-
tein, transthyretin. Gen Comp Endocrinol
2000;​119:​241-55.
23. van Bennekum AM, Wei S, Gamble
MV, et al. Biochemical basis for depressed
serum retinol levels in transthyretin-defi-
cient mice. J Biol Chem 2001;​276:​1107-
13.
24. Finn JD, Smith AR, Patel MC, et al.
A single administration of CRISPR/Cas9
lipid nanoparticles achieves robust and
persistent in vivo genome editing. Cell
Rep 2018;​22:​2227-35.
25. Wood K, Pink M, Seitzer J, et al. De-
velopment of NTLA-2001, a CRISPR/Cas9
genome editing therapeutic for the treat-
ment of ATTR. Presented at the Second
European Congress for ATTR Amyloido-
sis, Berlin, September 1–3, 2019.
26. Sabnis S, Kumarasinghe ES, Salerno
T, et al. A novel amino lipid series for
mRNA delivery: improved endosomal es-
cape and sustained pharmacology and
safety in non-human primates. Mol Ther
2018;​26:​1509-19.
27. Akinc A, Maier MA, Manoharan M,
et al. The Onpattro story and the clinical
translation of nanomedicines containing
nucleic acid-based drugs. Nat Nanotech-
nol 2019;​14:​1084-7.
28. Bae S, Park J, Kim JS. Cas-OFFinder:
a fast and versatile algorithm that search-
es for potential off-target sites of Cas9
RNA-guided endonucleases. Bioinformat-
ics 2014;​30:​1473-5.
29. Tsai SQ, Zheng Z, Nguyen NT, et al.
GUIDE-seq enables genome-wide profil-
ing of off-target cleavage by CRISPR-Cas
nucleases. Nat Biotechnol 2015;​33:​187-97.
30. Cameron P, Fuller CK, Donohoue PD,
et al. Mapping the genomic landscape of
CRISPR-Cas9 cleavage. Nat Methods 2017;​
14:​600-6.
31. Müller ML, Butler J, Heidecker B.
Emerging therapies in transthyretin am-
yloidosis — a new wave of hope after
years of stagnancy? Eur J Heart Fail 2020;​
22:​39-53.
32. Lin Y, Cradick TJ, Brown MT, et al.
CRISPR/Cas9 systems have off-target ac-
tivity with insertions or deletions between
target DNA and guide RNA sequences.
Nucleic Acids Res 2014;​42:​7473-85.
33. Haapaniemi E, Botla S, Persson J,
Schmierer B, Taipale J. CRISPR-Cas9 ge-
nome editing induces a p53-mediated DNA
damage response. Nat Med 2018;​24:​927-30.
34. Enache OM, Rendo V, Abdusamad M,
et al. Cas9 activates the p53 pathway and
selects for p53-inactivating mutations. Nat
Genet 2020;​52:​662-8.
35. Vilenchik MM, Knudson AG. Endoge-
nous DNA double-strand breaks: produc-
tion, fidelity of repair, and induction of
cancer. Proc Natl Acad Sci U S A 2003;​100:​
12871-6.
36. Varga T, Aplan PD. Chromosomal ab-
errations induced by double strand DNA
breaks. DNA Repair (Amst) 2005;​4:​1038-46.
Copyright © 2021 Massachusetts Medical Society.

502 n engl j med 385;6  nejm.org  August 5, 2021

The New England Journal of Medicine

Downloaded from nejm.org on July 11, 2023. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.